Trying To Compare Your Analytics Data Post GA4 Transition? Here’s Why It Doesn’t Add Up
After having set up Google Analytics 4, many website owners or marketing teams have tried to compare their historical data in GA4 with their data in Universal Analytics properties. However, there are metrics that allow comparison and some that don’t, which is leading to much confusion.
In this article, I go over the main metrics and how they are calculated in both versions, so that you can understand what you can and can’t compare. After reading this article, you will have a deeper understanding on the differences between the two GA versions, so that you can adjust reporting and include clear explanations for the wider team.
The very first instance of Google Analytics was launched in 2005, and Universal Analytics replaced it in 2012. This became the default property type and was complemented by Analytics 360 in 2016.
It’s no surprise then, that a further update was needed – and on 1st July 2023, Universal Analytics was replaced by GA4. Actually, GA4 has existed in one form or another since October 2020 but is now the full replacement for UA and so any organization not using it exclusively previously will now have moved across.
Both UA and GA4 strive to supply marketing teams, as well as webmasters and website owners with useful insights and intelligence on their website’s user behaviour, traffic data and commercials. However, the way in which these metrics are measured and provided differ between the two systems – so how and if these can be compared depends on the calculations behind them.
While Universal Analytics had two User metrics (Total Users and New Users), GA4 presents a third in addition to these: Active Users.
While Total Users previously referred to as simply the total number of users, in GA4 it is defined as the total number of unique users who logged an event. New Users has also been enhanced – where before it equated to the number of users who interacted with your site for the first time, it now also includes launching your app for the first time. This is measured via the number of new unique user IDs that logged the ‘first_open’ or ‘first_visit’ event.
Active Users, the new metric, is the number of distinct users who visited either your website or app. This is measured as any user who has an engaged session or where the software collects the ‘first_visit’ or ‘engagement_time_msec’ parameter on a website, ‘first_open’ or ‘engagement_time_msec’ on an Android app, or ‘first_open’ or ‘user_engagement’ from an iOS app.
The additional metric here does shift the onus somewhat in User reporting across Analytics. Whereas Universal Analytics tended to focus on Total Users across most reports, this will now be replaced by Active Users as a more useful and comprehensive measurement. However, Total Users from Universal Analytics can be compared to the Total Users in GA4 using the Explorations tab – but care should be taken as the definitions do differ slightly.
While the overarching definition of Pageviews remains the same between UA and GA4, it is known simply as Views on GA4. Screenviews in the previous version was measured in separate mobile-specific properties but both web and app data is now presented in the same property. This means that when tracking both web and app data, the additional app traffic is considered.
Previously, Universal Analytics allowed for additional filtering during data set comparisons. Such filtering is not supported in GA4, which means that not only Pageview count differs widely between both, but also the UA reporting may be subject to exclusion filters that the newer figures do not have. So, if you wish to compare ‘apples to apples’, it is important to put the same filters in place across both systems.
In UA, sessions referred to the period of time a user was actively engaged with a website or app with defined parameters as to what would incur the end of a session. In GA4, this has been redefined somewhat and renamed as Session Starts.
Now, in GA4, the ‘session_start’ event generates a session ID, which Analytics is able to associate with all subsequent events within that session. Sessions are no longer restarted at midnight or when new campaign parameters are encountered and will end whenever a period of more than 30-mins of inactivity occurs.
The changes in Session metric calculations does mean that the count may vary between UA and GA4 dependent on several factors, such as: geography (as timezones will impact when midnight resets sessions in UA), the use of UTMs on websites or apps (this too resets sessions on UA but not GA4), and filters that may exclude data. What’s more, bear in mind that GA4 uses a statistical estimate of the number of sessions using unique session IDs, while UA does not. This is more efficient and provides higher accuracy but can make direct comparisons pretty much impossible.
Conversions can be a fairly close metric for comparison between Universal Analytics and GA4, if based on destination URLs or on UA events. However, there are some differences that may impact comparisons if the measurements are a bit more complex.
In UA, a conversion goal was indicated as a particular user action being completed (destination, duration, pages/session, smart goals and event goals) – with one Conversion per session, per goal. In GA4, a Conversion Event is specific for each action stipulated, and counts event instance of that event, even if completed more than once during a session.
This is important: if direct comparisons between UA and GA4 are required for Conversions, it is recommended that the GA4 Conversion counting method setting is updated to ‘once per session’ for better accuracy.
Bounce Rate in Universal Analytics was calculated based on the percentage of single page sessions with no interactions within the page and a duration of 0 seconds. Now, in GA4, this refers to the percentage of sessions that are not engaged; so perhaps lasting longer than 0 seconds but leaving without having triggered any other events or visiting any other pages – which can be considered the opposite of engagement rate.
The older calculation has long been considered somewhat outdated, as there may be a myriad of reasons for users to visit a single-page application and leave without triggering an event – for example, reading a blog before leaving. Now, with a more comprehensive calculation being completed to measure Bounce Rate, the metric is more appropriate for modern website usage and can be considered a much better indicator of accuracy. This does, however, make comparisons between the two systems near impossible as the calculations vary so much.
The Last Word Google continues to innovate and develop through its webmaster offerings, and GA4 truly hones its metrics to a degree of accuracy not seen before. While comparing the system to its predecessor may be tricky for a while, it is clear that users will soon recognise the benefit of an updated and robust set of analytics – and that their digital approaches will only improve as a result.
I am a digital marketing professional with over 25 years’ experience in the sector and a passion for the internet industry. I build and implement online marketing strategies and campaigns to achieve my clients’ goals across a wide variety of sectors. As a marketing leader, I use my expertise in search, content, social and lead generation strategies to consistently improve and shape commercial success that drive growth, engagement and business development.
